LASSO-Cox Modeling of Survival Using High-Resolution CT-Based Radiomic Features in a Cohort of COVID-19 Patients and Its Generalizability to Standard Image Reconstruction

Background: Few studies have focused on predicting the overall survival (OS) of patients affected by SARS-CoV-2 (i.e., COVID-19) using radiomic features (RFs) extracted from computer tomography (CT) images. Reconstruction of CT scans might potentially affect the values of RFs. Methods: Out of 435 patients, 239 had the scans reconstructed with a single modality, and hence, were used for training/testing, and 196 were reconstructed with two modalities were used as validation to evaluate RFs robustness to reconstruction. During training, the dataset was split into train/test using a 70/30 proportion, randomizing the procedure 100 times to obtain 100 different models. In all cases, RFs were normalized using the z-score and then given as input into a Cox proportional-hazards model regularized with the Least Absolute Shrinkage and Selection Operator (LASSO-Cox), used for feature selection and developing a robust model. The RFs retained multiple times in the models were also included in a final LASSO-Cox for developing the predictive model. Thus, we conducted sensitivity analysis increasing the number of retained RFs with an occurrence cut-off from 11% to 60%. The Bayesian information criterion (BIC) was used to identify the cut-off to build the optimal model. Results: The best BIC value indicated 45% as the optimal occurrence cut-off, resulting in five RFs used for generating the final LASSO-Cox. All the Kaplan-Meier curves of training and validation datasets were statistically significant in identifying patients with good and poor prognoses, irrespective of CT reconstruction. Conclusions: The final LASSO-Cox model maintained its predictive ability for predicting the OS in COVID-19 patients irrespective of CT reconstruction algorithms.

Outcome Prediction for SARS-CoV-2 Patients Using Machine Learning Modeling of Clinical, Radiological, and Radiomic Features Derived from Chest CT Images

(1) Background: Chest Computed Tomography (CT) has been proposed as a non-invasive method for confirming the diagnosis of SARS-CoV-2 patients using radiomic features (RFs) and baseline clinical data. The performance of Machine Learning (ML) methods using RFs derived from semi-automatically segmented lungs in chest CT images was investigated regarding the ability to predict the mortality of SARS-CoV-2 patients. (2) Methods: A total of 179 RFs extracted from 436 chest CT images of SARS-CoV-2 patients, and 8 clinical and 6 radiological variables, were used to train and evaluate three ML methods (Least Absolute Shrinkage and Selection Operator [LASSO] regularized regression, Random Forest Classifier [RFC], and the Fully connected Neural Network [FcNN]) for their ability to predict mortality using the Area Under the Curve (AUC) of Receiver Operator characteristic (ROC) Curves. These three groups of variables were used separately and together as input for constructing and comparing the final performance of ML models. (3) Results: All the ML models using only RFs achieved an informative level regarding predictive ability, outperforming radiological assessment, without however reaching the performance obtained with ML based on clinical variables. The LASSO regularized regression and the FcNN performed equally, both being superior to the RFC. (4) Conclusions: Radiomic features based on semi-automatically segmented CT images and ML approaches can aid in identifying patients with a high risk of mortality, allowing a fast, objective, and generalizable method for improving prognostic assessment by providing a second expert opinion that outperforms human evaluation.

Dose-Effects Models for Space Radiobiology: An Overview on Dose-Effect Relationships.

Space radiobiology is an interdisciplinary science that examines the biological effects of ionizing radiation on humans involved in aerospace missions. The dose-effect models are one of the relevant topics of space radiobiology. Their knowledge is crucial for optimizing radioprotection strategies (e.g., spaceship and lunar space station-shielding and lunar/Mars village design), the risk assessment of the health hazard related to human space exploration, and reducing damages induced to astronauts from galactic cosmic radiation. Dose-effect relationships describe the observed damages to normal tissues or cancer induction during and after space flights. They are developed for the various dose ranges and radiation qualities characterizing the actual and the forecast space missions [International Space Station (ISS) and solar system exploration]. Based on a Pubmed search including 53 papers reporting the collected dose-effect relationships after space missions or in ground simulations, 7 significant dose-effect relationships (e.g., eye flashes, cataract, central nervous systems, cardiovascular disease, cancer, chromosomal aberrations, and biomarkers) have been identified. For each considered effect, the absorbed dose thresholds and the uncertainties/limitations of the developed relationships are summarized and discussed. The current knowledge on this topic can benefit from further in vitro and in vivo radiobiological studies, an accurate characterization of the quality of space radiation, and the numerous experimental dose-effects data derived from the experience in the clinical use of ionizing radiation for diagnostic or treatments with doses similar to those foreseen for the future space missions. The growing number of pooled studies could improve the prediction ability of dose-effect relationships for space exposure and reduce their uncertainty level. Novel research in the field is of paramount importance to reduce damages to astronauts from cosmic radiation before Beyond Low Earth Orbit exploration in the next future. The study aims at providing an overview of the published dose-effect relationships and illustrates novel perspectives to inspire future research.

COVID-19 Pandemic-Adapted Radiotherapy Guidelines: Are They Really Followed?

BACKGROUND: In our department, we provided guidelines to the radiation oncologists (ROs) regarding the omission, delay, or shortening of radiotherapy (RT). The purpose was to reduce the patients' exposure to the hospital environment and to minimize the departmental overcrowding. The aim was to evaluate the ROs' compliance to these guidelines. METHODS: ROs were asked to fill out a data collection form during patients' first visits in May and June 2020. The collected data included the ROs' age and gender, patient age and residence, RT purpose, treated tumor, the dose and fractionation that would have been prescribed, and RT changes. The chi-square test and binomial logistic regression were used to analyze the correlation between the treatment prescription and the collected parameters. RESULTS: One hundred and twenty-six out of 205 prescribed treatments were included in this analysis. Treatment was modified in 61.1% of cases. More specifically, the treatment was omitted, delayed, or shortened in 7.9, 15.9, and 37.3% of patients, respectively. The number of delivered fractions was reduced by 27.9%. A statistically significant correlation (p = 0.028) between younger patients' age and lower treatment modifications rate was recorded. CONCLUSION: Our analysis showed a reasonably high compliance of ROs to the pandemic-adapted guidelines. The adopted strategy was effective in reducing the number of admissions to our department.

Classification Performance for COVID Patient Prognosis from Automatic AI Segmentation—A Single-Center Study

Background: COVID assessment can be performed using the recently developed individual risk score (prediction of severe respiratory failure in hospitalized patients with SARS-COV2 infection, PREDI-CO score) based on High Resolution Computed Tomography. In this study, we evaluated the possibility of automatizing this estimation using semi-supervised AI-based Radiomics, leveraging the possibility of performing non-supervised segmentation of ground-glass areas. Methods: We collected 92 from patients treated in the IRCCS Sant’Orsola-Malpighi Policlinic and public databases;each lung was segmented using a pre-trained AI method;ground-glass opacity was identified using a novel, non-supervised approach;radiomic measurements were collected and used to predict clinically relevant scores, with particular focus on mortality and the PREDI-CO score. We compared the prediction obtained through different machine learning approaches. Results: All the methods obtained a well-balanced accuracy (70%) on the PREDI-CO score but did not obtain satisfying results on other clinical characteristics due to unbalance between the classes. Conclusions: Semi-supervised segmentation, implemented using a combination of non-supervised segmentation and feature extraction, seems to be a viable approach for patient stratification and could be leveraged to train more complex models. This would be useful in a high-demand situation similar to the current pandemic to support gold-standard segmentation for AI training.

Visceral fat shows the strongest association with the need of intensive care in patients with COVID-19.

BACKGROUND: Obesity was recently identified as a major risk factor for worse COVID-19 severity, especially among the young. The reason why its impact seems to be less pronounced in the elderly may be due to the concomitant presence of other comorbidities. However, all reports only focus on BMI, an indirect marker of body fat. AIM: To explore the impact on COVID-19 severity of abdominal fat as a marker of body composition easily collected in patients undergoing a chest CT scan. METHODS: Patients included in this retrospective study were consecutively enrolled among those admitted to an Emergency Department in Rome, Italy, who tested positive for SARS-Cov-2 and underwent a chest CT scan in March 2020. Data were extracted from electronic medical records. RESULTS: 150 patients were included (64.7% male, mean age 64 ±â€¯16 years). Visceral fat (VAT) was significantly higher in patients requiring intensive care (p = 0.032), together with age (p = 0.009), inflammation markers CRP and LDH (p < 0.0001, p = 0.003, respectively), and interstitial pneumonia severity as assessed by a Lung Severity Score (LSS) (p < 0.0001). Increasing age, lymphocytes, CRP, LDH, D-Dimer, LSS, total abdominal fat as well as VAT were found to have a significant univariate association with the need of intensive care. A multivariate analysis showed that LSS and VAT were independently associated with the need of intensive care (OR: 1.262; 95%CI: 1.0171-1.488; p = 0.005 and OR: 2.474; 95%CI: 1.017-6.019; p = 0.046, respectively). CONCLUSIONS: VAT is a marker of worse clinical outcomes in patients with COVID-19. Given the exploratory nature of our study, further investigation is needed to confirm our findings and elucidate the mechanisms underlying such association.